Skip to contentSkip to conversation

security: Yama LSM

PATCH Superseded Kees Cook 3 participants 6 messages Jun 23, 2010

Subsystems (4)

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

Kees Cook

author
#0 lore raw

PATCH v2 1/1 security: Yama LSM

This adds the Yama Linux Security Module to collect several security features (symlink, hardlink, and PTRACE restrictions) that have existed in various forms over the years and have been carried outside the mainline kernel by other Linux distributions like Openwall and grsecurity.

---
v2:
- add rcu locking, thanks to Tetsuo Handa.
- add Documentation/Yama.txt for summary of features.

---
 Documentation/Yama.txt   |   91 +++++++++++++++++
 security/Kconfig         |    6 +
 security/Makefile        |    2 +
 security/yama/Kconfig    |   72 ++++++++++++++
 security/yama/Makefile   |    3 +
 security/yama/yama_lsm.c |  240 ++++++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 414 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/Yama.txt
 create mode 100644 security/yama/Kconfig
 create mode 100644 security/yama/Makefile
 create mode 100644 security/yama/yama_lsm.c
Documentation/Yama.txt new file +91 -0
new file mode 100644
index 0000000..f9f15d2
@@ -0,0 +1,91 @@
+Yama is a Linux Security Module that collects a number of security
+protections that are not handled by the core kernel itself.  To select
+it at boot time, specify "security=yama" (though this will disable any
+other LSM).
+
+Yama is controlled through sysctl in /proc/sys/kernel/yama:
+
+- protected_sticky_symlinks
+- protected_nonaccess_hardlinks
+- ptrace_scope
+
+==============================================================
+
+protected_sticky_symlinks:
+
+A long-standing class of security issues is the symlink-based
+time-of-check-time-of-use race, most commonly seen in world-writable
+directories like /tmp. The common method of exploitation of this flaw
+is to cross privilege boundaries when following a given symlink (i.e. a
+root process follows a symlink belonging to another user).  For a likely
+incomplete list of hundreds of examples across the years, please see:
+http://cve.mitre.org/cgi-bin/cvekey.cgi?keyword=/tmp
+
+When set to "0", symlink following behavior is unrestricted.
+
+When set to "1" symlinks are permitted to be followed only when outside
+a sticky world-writable directory, or when the uid of the symlink and
+follower match, or when the directory owner matches the symlink's owner.
+
+This protection is based on the restrictions in Openwall and grsecurity.
+
+==============================================================
+
+protected_nonaccess_hardlinks:
+
+Hardlinks can be abused in a similar fashion to symlinks in sticky
+world-writable directories, but their weakness is not limited to
+just that scenario. For example, if /etc and /home are on the same
+partition, a regular user can create a hardlink to /etc/shadow in their
+home directory. While it retains the original owner and permissions,
+it is possible for privileged programs that are otherwise symlink-safe
+to mistakenly access the file through its hardlink. Additionally, a very
+minor untraceable quota-bypassing local denial of service is possible by
+an attacker exhausting disk space by filling a world-writable directory
+with hardlinks.
+
+When set to "0", hardlink creation behavior is unrestricted.
+
+When set to "1", hardlinks cannot be created to files that a given user
+would be unable to read and write originally, or are otherwise sensitive.
+
+This protection is based on the restrictions in Openwall and grsecurity.
+
+==============================================================
+
+ptrace_scope:
+
+As Linux grows in popularity, it will become a larger target for
+malware. One particularly troubling weakness of the Linux process
+interfaces is that a single user is able to examine the memory and
+running state of any of their processes. For example, if one application
+(e.g. Pidgin) was compromised, it would be possible for an attacker to
+attach to other running processes (e.g. Firefox, SSH sessions, GPG agent,
+etc) to extract additional credentials and continue to expand the scope
+of their attack without resorting to user-assisted phishing.
+
+This is not a theoretical problem. SSH session hijacking
+(http://www.storm.net.nz/projects/7) and arbitrary code injection
+(http://c-skills.blogspot.com/2007/05/injectso.html) attacks already
+exist and remain possible if PTRACE is allowed to operate as before.
+PTRACE is not commonly used by non-developers and non-admins, so system
+builders should be allowed the option to disable this debugging system.
+
+For a solution, some applications use prctl(PR_SET_DUMPABLE, ...) to
+specifically disallow such PTRACE attachment (e.g. ssh-agent), but many
+do not. A more general solution is to only allow PTRACE directly from a
+parent to a child process (i.e. direct "gdb EXE" and "strace EXE" still
+work), or with CAP_SYS_PTRACE (i.e. "gdb --pid=PID", and "strace -p PID"
+still work as root).
+
+0 - classic PTRACE permissions: a process can PTRACE any other process
+    running under the same uid, as long as it is dumpable (i.e. did not
+    transition uids, start privileged, or have prctl(PR_SET_DUMPABLE...)
+    called).
+
+1 - child-only PTRACE: a process can PTRACE only its descendants when
+    the above classic criteria is also met.
+
+This protection is based on the restrictions in grsecurity.
+
+==============================================================
security/Kconfig +6 -0
index 226b955..0e3a5ac 100644
@@ -140,6 +140,7 @@ config LSM_MMAP_MIN_ADDR
 source security/selinux/Kconfig
 source security/smack/Kconfig
 source security/tomoyo/Kconfig
+source security/yama/Kconfig
 
 source security/integrity/ima/Kconfig
 
@@ -148,6 +149,7 @@ choice
 	default DEFAULT_SECURITY_SELINUX if SECURITY_SELINUX
 	default DEFAULT_SECURITY_SMACK if SECURITY_SMACK
 	default DEFAULT_SECURITY_TOMOYO if SECURITY_TOMOYO
+	default DEFAULT_SECURITY_YAMA if SECURITY_YAMA
 	default DEFAULT_SECURITY_DAC
 
 	help
@@ -163,6 +165,9 @@ choice
 	config DEFAULT_SECURITY_TOMOYO
 		bool "TOMOYO" if SECURITY_TOMOYO=y
 
+	config DEFAULT_SECURITY_YAMA
+		bool "Yama" if SECURITY_YAMA=y
+
 	config DEFAULT_SECURITY_DAC
 		bool "Unix Discretionary Access Controls"
 
@@ -173,6 +178,7 @@ config DEFAULT_SECURITY
 	default "selinux" if DEFAULT_SECURITY_SELINUX
 	default "smack" if DEFAULT_SECURITY_SMACK
 	default "tomoyo" if DEFAULT_SECURITY_TOMOYO
+	default "yama" if DEFAULT_SECURITY_YAMA
 	default "" if DEFAULT_SECURITY_DAC
 
 endmenu
security/Makefile +2 -0
index da20a19..04354d1 100644
@@ -6,6 +6,7 @@ obj-$(CONFIG_KEYS)			+= keys/
 subdir-$(CONFIG_SECURITY_SELINUX)	+= selinux
 subdir-$(CONFIG_SECURITY_SMACK)		+= smack
 subdir-$(CONFIG_SECURITY_TOMOYO)        += tomoyo
+subdir-$(CONFIG_SECURITY_YAMA)		+= yama
 
 # always enable default capabilities
 obj-y					+= commoncap.o
@@ -19,6 +20,7 @@ obj-$(CONFIG_SECURITY_SELINUX)		+= selinux/built-in.o
 obj-$(CONFIG_SECURITY_SMACK)		+= smack/built-in.o
 obj-$(CONFIG_AUDIT)			+= lsm_audit.o
 obj-$(CONFIG_SECURITY_TOMOYO)		+= tomoyo/built-in.o
+obj-$(CONFIG_SECURITY_YAMA)		+= yama/built-in.o
 obj-$(CONFIG_CGROUP_DEVICE)		+= device_cgroup.o
 
 # Object integrity file lists
security/yama/Kconfig new file +72 -0

72 lines not on this page · raw message

security/yama/Makefile new file +3 -0
new file mode 100644
index 0000000..8b5e065
@@ -0,0 +1,3 @@
+obj-$(CONFIG_SECURITY_YAMA) := yama.o
+
+yama-y := yama_lsm.o
security/yama/yama_lsm.c new file +240 -0

240 lines not on this page · raw message

Signature · 6 lines
1.7.1


-- 
Kees Cook
Ubuntu Security Team
#1 lore raw
Quoted 4 lines · Kees Cook
security/yama/Kconfig new file +3 -0
+
+config SECURITY_YAMA_SYMLINKS
+	bool "Yama: protect symlink following in sticky world-writable

dirs"

IMHO it's bad style to have CONFIGs that just set defaults, if that can be done at runtime too. Especially as in your case if it's a lot of settings. Is it that bad to have a init script and drop these CONFIGs?

However the help texts are useful, these should be in the sysctl documentatin in Documentation instead.

security/yama/yama_lsm.c line 61

Quoted 5 lines
security/yama/yama_lsm.c new file +5 -0
+	if (rc) {
+		printk_ratelimited(KERN_INFO "ptrace of non-child"
+			" pid %d was attempted by: %s (pid %d)\n",
+			child->pid, get_task_comm(name, current),
+			current->pid);

It's probably obscure and other kernel code has this too, but at some point there were attacks to use terminal ESC sequences to attack root's terminal when they dmesg. Couldn't that be done through "comm" here?

At least the other code who has this problem doesn't claim to enhance security :)

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.
author
#2 lore raw

Hi Andi,

Quoted 10 lines · Andi Kleen
Quoted 4 lines · Kees Cook
security/yama/Kconfig new file +3 -0
+
+config SECURITY_YAMA_SYMLINKS
+	bool "Yama: protect symlink following in sticky world-writable

dirs"

IMHO it's bad style to have CONFIGs that just set defaults, if that can be done at runtime too. Especially as in your case if it's a lot of settings. Is it that bad to have a init script and drop these CONFIGs?

Oh, er, I actually added these configs because Eric Paris recommended them as handy for distributions. I'm fine with dropping them, but will I be asked to add them back later? And what about the case of CONFIG_SYSCTL being unset?

Quoted 2 lines

However the help texts are useful, these should be in the sysctl documentatin in Documentation instead.

I documented the sysctls in Documentation/Yama.txt (where SELinux.txt and Smack.txt live) should I create an additional file in Documentation/sysctl/ with that information (moved from Yama.txt) or move Yama.txt there?

Quoted 9 lines

security/yama/yama_lsm.c line 61

Quoted 5 lines
security/yama/yama_lsm.c new file +5 -0
+	if (rc) {
+		printk_ratelimited(KERN_INFO "ptrace of non-child"
+			" pid %d was attempted by: %s (pid %d)\n",
+			child->pid, get_task_comm(name, current),
+			current->pid);

It's probably obscure and other kernel code has this too, but at some point there were attacks to use terminal ESC sequences to attack root's terminal when they dmesg. Couldn't that be done through "comm" here?

I actually think this is a flaw in get_task_comm. (Though actually it's the fault of terminals if they process dangerous escape sequences. Worst case tends to just be confusing output, but that's not important -- nothing should spew non-printables regardless.) Would a patch to get_task_comm be accepted to replace non-printables with "?" or something when filling the buffer?

Thanks,

-Kees

Signature · 2 lines
Kees Cook
Ubuntu Security Team
#3 lore raw
Quoted 2 lines · Kees Cook

Oh, er, I actually added these configs because Eric Paris recommended them as handy for distributions. I'm fine with dropping them, but will I be

Distributions should use init scripts for that imho.

Quoted 2 lines

asked to add them back later? And what about the case of CONFIG_SYSCTL being unset?

"depends on SYSCTL"

Quoted 7 lines
Quoted 3 lines

However the help texts are useful, these should be in the sysctl documentatin in Documentation instead.

I documented the sysctls in Documentation/Yama.txt (where SELinux.txt and Smack.txt live) should I create an additional file in Documentation/sysctl/ with that information (moved from Yama.txt) or move Yama.txt there?

Ok missed that. what you prefer.

Quoted 6 lines

I actually think this is a flaw in get_task_comm. (Though actually it's the fault of terminals if they process dangerous escape sequences. Worst case tends to just be confusing output, but that's not important -- nothing should spew non-printables regardless.) Would a patch to get_task_comm be accepted to replace non-printables with "?" or something when filling the buffer?

Probably makes sense, although you might annoy someone who wants Kanji process names in UTF-8 or similar.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.
#4 lore raw
Quoted 9 lines
Quoted 6 lines

I actually think this is a flaw in get_task_comm. (Though actually it's the fault of terminals if they process dangerous escape sequences. Worst case tends to just be confusing output, but that's not important -- nothing should spew non-printables regardless.) Would a patch to get_task_comm be accepted to replace non-printables with "?" or something when filling the buffer?

Probably makes sense, although you might annoy someone who wants Kanji process names in UTF-8 or similar.

The task name is irrelevant, user controlled information. It's not really ideal for use in security logging.

Alan

#5 lore raw
Quoted 7 lines · Alan Cox
Quoted 7 lines
Quoted 6 lines

I actually think this is a flaw in get_task_comm. (Though actually it's the fault of terminals if they process dangerous escape sequences. Worst case tends to just be confusing output, but that's not important -- nothing should spew non-printables regardless.) Would a patch to get_task_comm be accepted to replace non-printables with "?" or something when filling the buffer?

Quoted 5 lines
Quoted 2 lines

Probably makes sense, although you might annoy someone who wants Kanji process names in UTF-8 or similar.

The task name is irrelevant, user controlled information. It's not really ideal for use in security logging.

Yes but the pid only is normally completely useless, because nobody knows afterwards what some old pid was.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.